EXTENSION ACROSS TEMPE TERRA, MARS FROM MOLA TOPOGRAPHIC MEASUREMENTS. B.W. Harrington1, R.J. Phillips1, and M.P. Golombek2, 1Washington University, Dept. Of Earth and Planetary Sciences, 1 Brookings Drive, St. Louis, MO 63130, [email protected], 2Jet Propulsion Lab, Califor- nia Institute of Technology, Pasadena, CA.

Introduction: Previously, methods of estimating ters), narrow-width (a few kilometer) structures that the extension across grabens and rifts on Mars by ne- have floors downdropped tens to hundreds of meters. cessity relied on photogeologic methods such as Larger, complex grabens and rifts are also found on shadow measurement, crater ellipticity, or photocli- Mars. These structures are wider (up to 100 km wide) nometry [1],[2]. With the new data being returned by and deeper (up to a few kilometers deep) than simple the Mars Global Surveyor, specifically from the Mars grabens, with individual border faults that can have up Orbital Laser Altimeter (MOLA), it is now possible to to a kilometer or more of relief [4]. In Tempe Terra directly measure the depths of these structures and these larger rifts are less than 50 km wide and around therefore to estimate more accurately the amount of a kilometer deep. extension. Here we provide an example of this new A variety of mechanical arguments and direct ob- approach in the Tempe Terra region. servations of simple graben fault dips on the Moon Geologic Setting: Tempe Terra is located on the and Mars suggest faults bounding grabens on Mars northeastern flank of the Tharsis volcanotectonic dip about 60o [4]. We have assumed this value for the province. It is an ancient terrain of Noachian age faults bounding the simple graben, complex graben, which has been cratered and heavily fractured. The and rifts. along with a large uncertainty (-±15o) in our western margin is embayed by volcanic flows origi- knowledge of the actual fault dip. This uncertainty is nating from Tharsis and Alba Patera while the eastern based on observed variations in normal fault dip on margin is buried under probable alluvium from Kasei the Earth and the standard deviation of the measured Valles and other outflow channels. simple graben fault dips on Mars. The plateau is cut by a series of mostly north to Passes and Extension: We measured the depths northeast striking graben, which formed during two along three MOLA groundtracks (214, 233, and 250) main stages of tectonic deformation. The older series that strike N10oE and intersect most of the structures of graben, which formed during the Middle to Late at and angle of 40 degrees from perpendicular. The Noachian, consist of both narrow simple graben and location of these groundtracks are shown in Figure 1, wider, deeper rifts and complex graben. In places along with the location of the two traverses used by these structures curve around a volcanic center located Golombek et. al. [5]. near 37oN, 77oW. The younger, mainly Late Hes- perian set of faults, strike slightly more to the west and cut across the older graben in places [3]. Extension From MOLA Depths: If the vertical structural relief (or throw), d, across a normal fault can be determinedand there is some knowledge of its initial dip, a, then it is possible to calculate the amount of extension or horizontal slip, Ex, between upthrown (footwall) and downthrown (hanging wall) blocks since

Ex = d / tan a To a first approximation, the depth of the structure is the same as the throw since most of the extensional fault scarps on Mars are sharply delineated with fairly flat footwall and hanging wall blocks. Of course, this ignores any subsequent deposition or infilling of the structure. Most extensional structures on Mars are simple Figure 1. Tempe Terra and the locations of the three MOLA grabens, which are bounded by two inward dipping groundtracks used (white lines) as well and the two traverses used in normal faults, with flat floors and symmetric scarps. [5] (black lines). They are typically long (tens to hundreds of kilome- Pass 250 begins at (30.0oN, 273.8oE) intersects the EXTENSION ACROSS TEMPE TERRA: B.W. Harrington, R.J. Phillips, and M.P. Golombek. southern traverse of Golombek, et al., [5] and ends at those assumed in Golombek et al. [5], , who used (43.0oN, 276.0oE). Passes 214 and 233 both cut across these slopes to estimate d values. The MOLA slope the northern traverse of Golombek, et al.., [5]. Pass distribution is shown in Figure 2. 214 begins at (33.0oN, 286.5oE) and ends at (48.0oN, 289.1oE) while pass 233 begins at (34.0oN, 290.0oE) and ends at (50.0oN, 292.8oE). The depths were meas- ured for any structures with fairly flat floors and symmetric fault scarps as well as those with asymmet- ric fault scarps when they could be identified from the Viking MDIM’s. In all cases, an attempt was made to take mass wasting into account when choosing likely candidates and measuring scarp depths. [This sen- tence seems to run on. Do you take mass wasting into account only for asymmetric fault scarps. Where pos- sible an attempt was made to correlate the structures from the MOLA groundtrack with available images of the area. However, the resolution of most of the im- ages was too poor to resolve some of the finer struc- ture readily seen in the MOLA groundtracks. Along Pass 214, a total of 59 scarps were meas- ured. The sum of the depths (12.0 km) indicates 6.9 Figure 2. A plot of all the measured slopes for both graben and rift km of extension, which results in only 0.67% strain. scarps. The majority of the extension occurred between 400 km and 600 km along the groundtrack where it Analysis of the strain across Tempe Terra can be crossed an intensely faulted region. compared to that predicted by various models of Thar- A total of 63 scarp depths were measured along sis formation and evolution. Ongoing work is exam- Pass 233, with a total depth of 9.0 km and a resulting ining the strain on Mars at a global scale and looking 5.2 km of extension or only 0.47% strain. The exten- at the strain around Tharsis as a function of geologic sion along this pass is fairly linear but does jump age. This allows a determination of the spatial strain slightly around 750 km where it crosses the same in- history of the Tharsis tectonic province as a whole. tensely fractured terrain that pass 214 did farther to References: the north. [1] Tanaka, K.L., and Davis, P.A. (1988), JGR, Pass 250 contained a total of 65 scarps. The total 93, 14,893-14,917. [2] Plescia, J.B., (1991), JGR, 96, depth was 10.0 km and the extension was 5.8 km to- 18,883-18,895. [3] Scott, D.H., and Tanaka, K.L, tal. This results in 0.64% strain along the ground- (1986), U.S. Geol. Surv. Misc. Inv. Ser. Map. I-1802- track. Pass 250 contains several regions of high strain 4. Scale 1:15,000,000. [4] Banerdt, W.B., Golombek, which correspond to three narrow, highly faulted re- M.P., and Tanaka, K.L., (1992), Stress and Tectonics gions. Thus, while the overall strain is small in many on Mars, Chapter 8, in MARS, edited by H.H.Keifer, places it can be locally as high as 10% to 15%. B.M Jakosky, C.W. Snyder, and M.S. Matthews, pp. Discussion: The strains along the groundtracks 249-297, Univ. of Ariz. Press, Tuscon. [5] Golombek, (0.67%, 0.47% and 0.64%) are significantly lower M.P., Tanaka, K.L., and Franklin B.J., (1996), JGR, than Golombek et al. [5] who estimated strains of 101, 26,119-26,130. 2.9±2.1% and 1.7±1.3% along the southern and northern traverses respectively, as well as 1.96% from an estimate of crater elongation. Differences in the results stem from two main sources. First, many of the structures measured by Golombek et al., [5] were par- tially buried or even absent in the regions of the MOLA groundtrack, thus reducing the estimated strain. This problem may be circumvented by the denser MOLA coverage now becoming available. Sec- ondly, the scarp slopes (especially those of the rifts) measured by MOLA are systematically lower than